Cognitive impairments among patients with substance use disorders are prevalent and associated with adverse treatment outcomes. However, knowledge of the predictive value of broad cognitive screening instruments on long-term treatment outcomes is limited. The present study aimed to examine the predictive value of measures from the Montreal Cognitive Assessment®(MoCA®), Wechsler Abbreviated Scale of Intelligence (WASI), and the Behaviour Rating Inventory of Executive Function–Adult version (BRIEF-A) on self-reported long-term substance use and abstinence in patients with polysubstance use disorders (pSUD). A cohort (N= 164) of patients with pSUD who started a new treatment sequence in the Stavanger University Hospital catchment area were recruited and followed prospectively for 5 years. Participants completed neurocognitive testing with the MoCA®, WASI, and BRIEF-A at inclusion and were categorized as cognitively impaired or non-impaired according to recommended cut-off values. The sum score of the items from the Drug Use Disorders Identification Test Consumption scale (DUDIT-C) was used as a measure of substance use outcome 1 and 5 years after inclusion. We defined substance abstinence (DUDIT-C= 0) and heavy substance use (DUDIT-C≥ 7) to determine whether cognitive impairments measured by the respective instruments were associated with and could predict abstinence and heavy substance use 1 and 5 years after baseline. At the 1-year follow-up, 54% of the total sample reported total abstinence from substances. Conversely, 31% presented heavy substance use. At 5 years, 64% of the total sample reported abstinence from substances, while 25% presented heavy substance use. The results showed a statistically significant association between cognitive impairment defined from MoCA® and higher continuous scores on DUDIT-C at 1-year follow-up. There were no differences in substance abstinence or heavy substance use between patients with and without cognitive impairment at the 1-and 5-year follow-ups. Furthermore, cognitive impairment did not explain substance abstinence or heavy substance use at the 1-and 5-year follow-ups. Generally, individuals with pSUD may be burdened and lack psychosocial resources to such an extent that cognitive functioning plays a subordinate role in long-term recovery. The present study suggests that results on screening tools assessing broad cognitive domains at treatment initiation have limited clinical value in predicting long-term substance use outcomes. There is a need to establish clinically viable instruments to assess cognitive functions with well-established clinical and ecological validity in the SUD population.